Hi Istvan,

> Hi Viktor,
> 
> As I have o more argues and we are back to some original issues, I am
> stopping to write more on this thread.
> But an another case should be analyzed too: if we define HB_FM_STD_ALLOC,
> everything goes well, without crashes.

It's also a system-wide setting (which BTW has a large performance 
hit for most targets), but most importantly it's also just a random 
shot in the dark (AFAICS).

Until we find out what is the real problem, such solutions would only 
mask or delay the real problem, so it's nothing more than a hack (with 
bad side-effects).

Fixing HBQT problems by swapping such basic components like memory 
allocator or MT/ST mode HVM, moreover on a per platform/compiler 
basis, looks wrong to me. We should understand what's going on 
exactly and fix it at appropriate level.

Brgds,
Viktor

_______________________________________________
Harbour mailing list (attachment size limit: 40KB)
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour

Reply via email to